[
https://issues.apache.org/jira/browse/SOLR-11838?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16419681#comment-16419681
]
Christine Poerschke commented on SOLR-11838:
--------------------------------------------
I've been playing around some more with deeplearning4j and Solr streaming
expressions ...
new pull request
[deeplearning4j/pull/4876|https://github.com/deeplearning4j/deeplearning4j/pull/4876]
shares the results, proposing to add a {{DataSetIterator}} implementation
(tentatively named {{TupleStreamDataSetIterator}}) which uses a [streaming
expression|https://lucene.apache.org/solr/guide/7_2/streaming-expressions.html]
to fetch data from Solr and/or one of the sources (e.g. {{jdbc}}) supported as
a [stream
source|https://lucene.apache.org/solr/guide/7_2/stream-source-reference.html].
... this is not specific to Learning-To-Rank and I have no specific real use
case in mind as yet but would be curious to hear if anyone could think of
scenarios where something other than fields are part of the streaming
expression used in the iterator?
> explore supporting Deeplearning4j NeuralNetwork models
> ------------------------------------------------------
>
> Key: SOLR-11838
> URL: https://issues.apache.org/jira/browse/SOLR-11838
> Project: Solr
> Issue Type: New Feature
> Reporter: Christine Poerschke
> Priority: Major
> Attachments: SOLR-11838.patch, SOLR-11838.patch
>
>
> [~yuyano] wrote in SOLR-11597:
> bq. ... If we think to apply this to more complex neural networks in the
> future, we will need to support layers ...
> [~malcorn_redhat] wrote in SOLR-11597:
> bq. ... In my opinion, if this is a route Solr eventually wants to go, I
> think a better strategy would be to just add a dependency on
> [Deeplearning4j|https://deeplearning4j.org/] ...
> Creating this ticket for the idea to be explored further (if anyone is
> interested in exploring it), complimentary to and independent of the
> SOLR-11597 RankNet related effort.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
